The bioenergetics of world class cycling

Professional cycle racing is one of the most demanding of all sports combining extremes of exercise duration, intensity and frequency. Riders are required to perform on a variety of surfaces (track, road, cross-country, mountain), terrains (level, uphill and downhill) and race situations (criterions, sprints, time trials, mass-start road races) in events ranging in duration from 10 s to 3 wk stage races covering 200 m to 4,000 km. Furthermore, professional road cyclists typically have approximately 100 race d/yr. Because of the diversity of cycle races, there are vastly different physiological demands associated with the various events. Until recently there was little information on the demands of professional cycling during training or competition. However, with the advent of reliable, valid bicycle crank dynanometers, it is now possible to quantify real-time power output, cadence and speed during a variety of track and road cycling races. This article provides novel data on the physiological demands of professional and world-class amateur cyclists and characterises some of the physiological attributes necessary for success in cycling at the élite level.     

Road gradient and cycling power: An observational study in male professional cyclists

ObjectivesTo investigate the influence of road gradient on cycling power output in male professional cyclists, and to determine whether cyclist typology (i.e., flat or climbing specialist) moderates this influence.DesignObservational study.MethodsNinety-eight professional cyclists (27 ± 6 years; 53 flat and 45 climbing specialists). We collected power output data during both training sessions and competitions over 10 years (2013–2022). We determined the maximal mean power values attained for efforts lasting 1, 5, 10 and 20 min, during both level cycling and uphill cycling (average slope< or ≥5 %, respectively), as well as the average road gradients on which cyclists attained their maximal mean power.ResultsMaximal mean power values were higher during uphill cycling than during level cycling for all effort durations (difference ranging between 0.4 and 3.6 %, all p < 0.003). This finding was confirmed for flat and uphill specialists separately (p < 0.003 for both), with a similar increase in maximal mean power values between level cycling and uphill cycling in the two typologies except for longer efforts (≥10 min), in which maximal mean power values tended to increase more in climbers. Participants attained maximal mean power at an average slope of 6.0–7.3 %, with no differences between effort durations or cyclist typologies.ConclusionsProfessional cyclists attain higher maximal mean power values on steep than on level road gradients regardless of their typology, with an average gradient of 6–7 % appearing optimal (or at least the most common) for achieving the highest maximal mean power values.     

Improvements in multi-joint leg function following chronic eccentric exercise

Previous authors have reported that chronic eccentric cycling facilitates greater changes in multi-joint leg function (hopping frequency, maximum jumping height) compared with concentric cycling. Our purpose was to evaluate changes in leg spring stiffness and maximum power following eccentric and concentric cycling training. Twelve individuals performed either eccentric (n=6) or concentric (n=6) cycling for 7 weeks (3 sessions/week) while training duration progressively increased. Participants performed trials of submaximal hopping, maximal counter movement jumps, and maximal concentric cycling to evaluate leg spring stiffness, maximum jumping power, and maximum concentric cycling power respectively, before and 1 week following training. Total work during training did not differ between eccentric and concentric cycling (126 ± 15-728 ± 91 kJ vs 125 ± 10-787 ± 76 kJ). Following training, eccentric cycling exhibited greater changes in k(leg) and jumping P(max) compared with CON(cyc) (10 ± 3% vs -2 ± 4% and 7 ± 2% vs -2 ± 3%, respectively, P=0.05). Alterations in CON(cyc) P(max) did not differ between ECC(cyc) (1035 ± 142 vs 1030 ± 133 W) and CON(cyc) (1072 ± 98 vs 1081 ± 85 W). These data demonstrate that eccentric cycling is an effective method for improving leg spring stiffness and maximum power during multi-joint tasks that include stretch-shortening cycles. Improvements in leg spring stiffness and maximum power would be beneficial for both aging and athletic populations.     

Torque and power-velocity relationships in cycling: relevance to track sprint performance in world-class cyclists

The aims of the present study were both to describe anthropometrics and cycling power-velocity characteristics in top-level track sprinters, and to test the hypothesis that these variables would represent interesting predictors of the 200 m track sprint cycling performance. Twelve elite cyclists volunteered to perform a torque-velocity test on a calibrated cycle ergometer, after the measurement of their lean leg volume (LLV) and frontal surface area (A(p)), in order to draw torque- and power-velocity relationships, and to evaluate the maximal power (P(max)), and both the optimal pedalling rate (f(opt)) and torque (T(opt)) at which P (max) is reached. The 200 m performances--i.e. velocity (V200) and pedalling rate (f 200)--were measured during international events (REC) and in the 2002 French Track Cycling Championships (NAT). P(max), f(opt), and T(opt) were respectively 1600 +/- 116 W, 129.8 +/- 4.7 rpm and 118.5 +/- 9.8 N . m. P(max) was strongly correlated with T(opt) (p < 0.001), which was correlated with LLV (p < 0.01). V200 was related to P(max) normalized by A(p) (p < or = 0.05) and also to f(opt) (p < 0.01) for REC and NAT. f 200 (155.2 +/- 3, REC; 149 +/- 4.3, NAT) were significantly higher than f(opt) (p < 0.001). These findings demonstrated that, in this population of world-class track cyclists, the optimization of the ratio between P(max) and A(p) represents a key factor of 200 m performance. Concerning the major role also played by f(opt), it is assumed that, considering high values of f 200, sprinters with a high value of optimal pedalling rate (i.e. lower f200-f(opt) difference) could be theoretically in better conditions to maximize their power output during the race and hence performance.     

The 4000-m team pursuit cycling world record: theoretical and practical aspects

Due to constant competition conditions, track cycling can be accurately modeled through physiological and biomechanical means. Mathematical modeling predicts an average workload of 520 W for every team member for a new team pursuit world record. Performance in team pursuit racing is highly dependent on aerobic capacity, anaerobic skills, and aerodynamic factors. The training concept of the 2000 record-breaking team pursuit team was based on unspecific training of these qualities and periodical, short-term recall of previously acquired track specific skills. Aerobic performance was trained through high overall training mileage (29,000-35,000 km.yr-1) with workload peaks during road stage races. Before major track events, anaerobic performance, and track-specific technical and motor skills were improved through discipline-specific track training. Training intensities were monitored through heart rate and lactate field tests during defined track-training bouts, based on previously performed laboratory exercise tests. During pursuit competition, analysis of half-lap split times allowed an estimation of the individual contribution of each rider to the team's performance and thereby facilitated modifications in team composition to optimize race speed. The theoretically predicted performance necessary for a new world record was achieved through careful planning of training and competition schedules based on a concise theoretical concept and the high physiological capacities of the participating athletes.     

Post-competition blood lactate concentrations in competitive track cyclists

Blood lactate (BL) concentrations were measured in 19 competitive cyclists after 4 competitive track events (kilometre, individual and team pursuit, and match sprints) during the US National Championships. A total of 34 samples were drawn. Our purpose was to investigate the share of anaerobic metabolism in the various events. The greatest mean BL concentration (16.94 mM/l) was found after the kilometre, with one individual recording a value of 18.22 mM/l. A low correlation coefficient for BL versus total riding time was found in the 6 samples drawn during the kilometre (r=0.51) and the 12 samples drawn during the individual pursuit (r=0.44). BL concentrations in all four events were not significantly different (p<0.05) from each other. High values for BL were found in the match sprints (7 samples, mean 13.65 mM/l) an event lasting approximately 11 seconds. The mean BL in the kilometre (6 samples, 16.94 mM/l), individual pursuit (12 samples, 15.18 mM/l), and team pursuit (5 samples, 12.08 mM/l) compare favourably to the values reported for a variety of exercise modes of similar time, except running. While there is increased muscle mass in running, this fact alone may not explain the lower BL values in cycling. It is suggested that lower BL levels may be attributed to prolonged contraction-relaxation phase during cycling causing decreased blood flow to the leg.     

Development of peak performance in track cycling.

BACKGROUND: Retrospective analysis of peak performances can be a usefull tool for the estimation of future trends in high performance sports. The purpose of this study was to investigate the evolution of performance in track cycling from 1979 to 1999 and to asses age- and gender-related differences. METHODS: We studied the results of the world track cycling championships for this period in 200 m, 1000 m, individual and team pursuit races for elite and junior athletes. Overall trends, performance differences between rank 1 and 5, gender- and age-related differences were calculated. RESULTS: They show a significant (p<0.01) improvement in 1000 m, individual and team pursuit times for men and individual pursuit times for women. No significant evolution was seen in 200 m performance. In junior riders, only male athletes showed a continuous, significant improvement of average race speed in individual and team pursuit over the study period. Gender difference was 11+/-1.8% in all disciplines at all ages. Difference between elite and junior riders ranged between 5+/-2.1% for male and 6.1+/-2.2% for female athletes. The gap between rank 1 and 5 remained constant (2-3%) over the study period. CONCLUSIONS: A continuous improvement of performance over the last 20 years is visible in track cycling endurance disciplines. New technical developments show no statistical significant impact. The performance gap between male and female athletes is constant, independent of discipline or race distance and comparable to observations in other sports. Age-related differences in performance is most visible in disciplines requiring short, high intensity power output. Based on these data, estimation of possible winning times and adaptation of training programs for future track cycling competitions might be facilitated.     

Maximising performance in triathlon: applied physiological and nutritional aspects of elite and non-elite competitions.

Triathlon is a sport consisting of sequential swimming, cycling and running. The main diversity within the sport of triathlon resides in the varying event distances, which creates specific technical, physiological and nutritional considerations for athlete and practitioner alike. The purpose of this article is to review physiological as well as nutritional aspects of triathlon and to make recommendations on ways to enhance performance. Aside from progressive conditioning and training, areas that have shown potential to improve triathlon performance include drafting when possible during both the swim and cycle phase, wearing a wetsuit, and selecting a lower cadence (60-80 rpm) in the final stages of the cycle phase. Adoption of a more even racing pace during cycling may optimise cycling performance and induce a "metabolic reserve" necessary for elevated running performance in longer distance triathlon events. In contrast, drafting in swimming and cycling may result a better tactical approach to increase overall performance in elite Olympic distance triathlons. Daily energy intake should be modified to reflect daily training demands to assist triathletes in achieving body weight and body composition targets. Carbohydrate loading strategies and within exercise carbohydrate intake should reflect the specific requirements of the triathlon event contested. Development of an individualised fluid plan based on previous fluid balance observations may assist to avoid both dehydration and hyponatremia during prolonged triathlon racing.     

Energy system contribution to 100-m and 200-m track running events.

While sprint track running events, lasting 10-25 secs, are characterised by an anaerobic metabolic dominance, no actual track running data exist which have quantified the relative energy system contributions. Using previous methods employed by our laboratory, including 'in race' measures of VO2, accumulated oxygen deficit (AOD), blood lactate concentration and estimated phosphocreatine degradation (La/PCr), the aerobic-anaerobic energy system contributions to 100-m and 200-m events were calculated. For the 100-m event, results indicated a relative aerobic-anaerobic energy system contribution (based on AOD measures) of 21%-79% and 25-75% for males and females respectively (9%-91% and 11%-89% based on La/PCr measures; p<0.05 for both genders for 100-m from AOD estimates). For the 200-m, a 28%-72% and 33%-67% contribution for male and female athletes was estimated (21%-79% and 22%-78% based on La/PCr measures; NS from AOD estimates). A range of energy system contribution estimates for events of these durations have previously been proposed using a variety of techniques. The data from the current study also show different results depending on the measurement technique utilised. While AOD measures are often used to estimate anaerobic energy contribution, at such high exercise intensities (and brief exercise durations) as used in the present study, AOD measures showed larger aerobic energy estimates than expected.     

The effect of age and gender on heart rate variability after endurance training

PURPOSE: This research investigated the age and gender differences in cardiovascular adaptation to a standardized/quantified endurance-training program that included two taper periods. METHODS: The latter was analyzed from spectral analysis of electrocardiogram records of heart rate variability (HRV) at rest in groups of young (19-21 yr) and middle aged (40-45 yr), mixed gender groups (6 males and 6 females), pre- and poststandardized training. All subjects were recreational runners who completed the same 12-wk running program. Before, and subsequent to training, HRV was measured during supine rest and submaximal cycling. RESULTS: There was a significant decrease in heart rate both at rest (2.7 +/- 0.45 beats x min-1) and during submaximal exercise (8.1 +/- 0.67 beats x min-1) in both age groups after training. After training, total spectral power increased (560.7 +/- 308.9 ms2), as well as high-frequency power (362.3 +/- 405.5 ms2), in both age groups at rest. The young group showed a greater increase in total power (849.0 +/- 308.7 ms2) after the training program. CONCLUSION: It is concluded that a well-designed 12-wk endurance-training program will decrease resting and submaximal heart rate in both younger and older adults. The significant increase in HRV, total power, and high-frequency power in all groups after endurance training indicates that HRV measurement appears to provide an effective, noninvasive assessment of cardiovascular adaptation to aerobic training.     

3周高住低练对女子自行车运动员无氧功的影响

目的:探讨3周高住低练(HiLo)对女子自行车运动员无氧能力的影响。方法:上海自行车队8名优秀女子运动员进行为期3周的HiLo,每天在低氧环境下(模拟海拔2500m)睡眠10个小时,每周6天,常氧下正常训练。分别于HiLo前后进行内容相同的10sec和60sec无氧功测试,测定运动后即刻、运动后恢复3min的心率。结果:(1)3周HiLo后,两种无氧功测试中的最大功率分别提高了3.43%和3.02%;(2)60sec无氧功测试中的平均功率显著提高了5.72%(P<0.05);(3)在60sec无氧功测试中,第1、2个20sec分段的平均功率分别提高了6.15%、6.17%(P<0.05)。结论:3周HiLo对女子自行车运动员的磷酸原供能能力未产生显著影响;运动员无氧糖酵解能力有所改善。     

Maximal leg-strength training improves cycling economy in previously untrained men

PURPOSE: This study examined cycling economy before and after 8 wk of maximal leg-strength training. METHODS: Seven previously untrained males (25 +/- 2 yr) performed leg-strength training 3 d.wk(-1) for 8 wk using four sets of five repetitions at 85% of one repetition maximum (1RM). Body mass, lean-leg muscle mass (LLM), percentage of body fat, and leg strength (1RM) were measured at 0, 4, and 8 wk of training. Cycling economy was calculated as the deltaVO2/deltaWR (change in the O2 cost of exercise divided by the change in the power between two different power outputs). RESULTS: There were significant increases in LLM and 1RM from 0 to 4 wk of training (LLM: 25.8 +/- 0.7 to 27.2 +/- 0.8 kg; 1RM: 138 +/- 9 to 215 +/- 9 kg). From 4 to 8 wk of training, 1RM continued to increase significantly (215 +/- 9 to 266 +/- 8 kg) with no further change observed in LLM. Peak power during incremental cycling increased significantly (305 +/- 14 to 315 +/- 16 W), whereas the power output achieved at the gas-exchange threshold (GET) remained unchanged. Peak O2 uptake and the O2 uptake achieved at the GET also remained unchanged following training. Cycling economy improved significantly when the power output was increased from below the GET to above the GET but not for power outputs below the GET. CONCLUSION: Maximal leg-strength training improves cycling economy in previously untrained subjects. Increases in leg strength during the final 4 wk of training with unchanged LLM suggest that neural adaptations were present.     

Muscle recruitment pattern in cycling: a review

Studies have indicated that the muscles work in a systematic and coordinated way to generate and direct power from the human body to the crank during cycling. Understanding of the muscle involvement or recruitment pattern during cycling will be useful for developing specific and effective muscle training and rehabilitation programs for cyclists. Moreover, it will also facilitate the use of the cycling ergometer for therapeutic purpose. This paper reviews the current literature on muscle recruitment pattern during cycling and the effects of muscle fatigue, cadence, riding posture and seat height on this recruitment pattern. In the power phase or ‘downstroke’, the hip, knee and ankle joints extend simultaneously for the pushing action, whilst in the recovery phase or ‘upstroke’, they flex together to pull the pedal back to the top dead center of the crank cycle. Recent studies have indicated that in this repeated sequence, the mono-articular muscles are mainly involved in the generation of positive work whereas the biarticular muscles are responsible for regulating force transmission. Some muscles co-activate during cycling to provide synergistic actions and other functional needs.Muscle fatigue is an important factor affecting cycling performance. It has been reported that muscle fatigue in the lower body would alter the cycling motion and muscle activation pattern. Therefore, studying the change of muscle activation pattern during cycling at the fatigued level may shed light on the sequel of local muscle fatigue. A muscle training program specifically for cycling can then be designed accordingly. Additionally, the change of cadence during cycling will affect the muscle recruitment pattern. There is a unique cadence that minimizes the muscle activation level at a specific level of power output. This cadence will increase as the power output increases. The change of riding posture from sitting to standing renders the pelvis unsupported and the body weight will assist the power phase of pedalling. Similarly, changes to the seat height will alter the posture which will affect the directions of muscle actions to the crank, thus changing the muscle recruitment pattern.     

Transfer effects of endurance training with the arms and legs

The purpose of this study was to examine whether endurance training in the form of arm or leg cycling resulted in significant transfer effects when exercise was performed with the untrained muscle group. Sixteen middle-aged male volunteers completed 24 training sessions over 8 wk on either an arm cycle ergometer (AG, N = 8, mean age = 35.2 +/- 6.6 yr) or a leg cycle ergometer (LG, N = 8, mean age = 41.0 +/- 4.7 yr). The two groups were initially equated for their pre-training peak oxygen uptake (pVO2) determined during leg cycling (44.5 +/- 5.0 and 43.8 +/- 7.7 ml.kg-1.min-1 for the AG and LG, respectively). Training was performed at an intensity that was mid-way between the pre-training ventilatory threshold (VT) and the pVO2 for both cycling methods. Significant increases (P less than .05) were observed in the relative values of the oxygen uptake at the VT and the pVO2 as a result of both these methods of training, but these elevations were specific to the muscle groups that were trained. This specificity of training was also evident in the cardiorespiratory and metabolic measurements obtained during submaximal steady state exercise performed at the power output corresponding to the pre-training VT during arm and leg exercise. Hence, it was concluded that improvements in exercise performance resulting from short-term aerobic training with the arms or legs in middle-aged males with relatively high aerobic powers are due primarily to peripheral adaptations in the trained muscles.     

Effects of plyometric training followed by a reduced training programme on physical performance in prepubescent soccer players

BACKGROUND: In adult population, stretch-shortening cycle exercise (plyometric exercise) is often used to improve leg muscle power and vertical jump performance. In children, limited information regarding this type of exercise is available. The purpose of this study was to examine the effectiveness of plyometric training and maintenance training on physical performances in prepubescent soccer players. METHODS: Twenty boys aged 12-13 years was divided in two groups (10 in each): jump group (JG) and control group (CG). JG trained 3 days/week during 10 weeks, and performed various plyometric exercises including jumping, hurdling and skipping. The subsequent reduced training period lasted 8 weeks. However, all subjects continued their soccer training. Maximal cycling power (Pmax) was calculated using a force-velocity cycling test. Jumping power was assessed by using the following tests: countermovement jump (CMJ), squat jump (SJ), drop jump (DJ), multiple 5 bounds (MB5) and repeated rebound jump for 15 seconds (RRJ15). Running velocities included: 20, 30 and 40 m (V20, V30, V40 m). Body fat percentage (BF percent) and lean leg volume were estimated by anthropometry. RESULTS: Before training, except for BF percent, all baseline anthropometric characteristics were similar between JG and CG. After the training programme, Pmax (p<0.01), CMJ (p<0.01), SJ (p<0.05), MB5 (p<0.01), RRJ15 (p<0.01) and V20 m (p<0.05), performances increased in the JG. During this period no significant performance increase was obtained in the CG. After the 8-week of reduced training, except Pmax (p<0.05) for CG, any increase was observed in both groups. CONCLUSIONS: These results demonstrate that short-term plyometric training programmes increase athletic performances in prepubescent boys. These improvements were maintained after a period of reduced training.     

Modeling sprint cycling using field-derived parameters and forward integration

We previously reported that a mathematical model could accurately predict steady-state road-cycling power when all the model parameters were known. Application of that model to competitive cycling has been limited by the need to obtain accurate parameter values, the non-steady-state nature of many cycling events, and because the validity of the model at maximal power has not been established. PURPOSE: We determined whether modeling parameters could be accurately determined during field trials and whether the model could accurately predict cycling speed during maximal acceleration using forward integration. METHODS: First, we quantified aerodynamic drag area of six cyclists using both wind tunnel and field trials allowing for these two techniques to be compared. Next, we determined the aerodynamic drag area of three world-class sprint cyclists using the field-test protocol. Track cyclists also performed maximal standing-start time trials, during which we recorded power and speed. Finally, we used forward integration to predict cycling speed from power-time data recorded during the maximal trials allowing us to compare predicted speed with measured speed. RESULTS: Field-based values of aerodynamic drag area (0.258 +/- 0.006 m) did not differ (P = 0.53) from those measured in a wind tunnel (0.261 +/- 0.006 m2). Forward integration modeling accurately predicted cycling speed (y = x, r2 = 0.989) over the duration of the standing-start sprints. CONCLUSIONS: Field-derived values for aerodynamic drag area can be equivalent to values derived from wind tunnel testing, and these values can be used to accurately predict speed even during maximal-power acceleration by world-class sprint cyclists. This model could be useful for assessing aerodynamic issues and for predicting how subtle changes in riding position, mass, or power output will influence cycling speed.     

Determination of optimal pacing strategy in track cycling with an energy flow model

The purpose of this study was to investigate the effect of pacing strategies on performance times in the 1000 m time trial event and the 4000 m pursuit event in track cycling. For this purpose, we simulated these events with a model based on the flow of energy in cycling. Different strategies in distributing the available anaerobic energy were evaluated and we compared model predictions of split times and final times with values achieved by cyclists during championships. The best result at the 1000 m time trial was obtained when the cyclist had the highest anaerobic peak power output and used an 'all-out' strategy. The fastest time on the 4000 m pursuit was achieved with an 'all-out' start at a high level of initial power output, followed by a constant anaerobic power output after 12 seconds, resulting in an evenly paced race. The results show that even small variations in pacing strategy may have substantial effects on performance. There seems to be an opportunity to gain a competitive advantage when individual athletes experiment with small variations in pacing strategy to find the precise individual strategy that works best under specific conditions.     

Improvement in maximal isokinetic cycle ergometry with cardiac rehabilitation

It is unclear whether improvements in short-term (30 s) exercise capacity are associated with the increased aerobic exercise tolerance frequently observed in cardiac patients following training. Carefully selected patients with documented coronary artery disease were randomly allocated either to a control group (N = 10) or to 12 wk of endurance exercise training (N = 12); both progressive incremental cycle ergometer testing (maximal power output and peak VO2) and 30 s maximal isokinetic cycle ergometry (peak power, total work, and fatigue index) were measured on entry into the study and 12 wk later. Initial maximum performance measures in progressive incremental exercise and in maximal short-term isokinetic cycling were similar in both groups. Following the training program, maximum power output measured during progressive incremental exercise increased by 21% (P less than 0.005) and peak VO2 increased by 18% (P less than 0.005) in the exercise group, but they were unchanged in the control group. Isokinetic peak power and total work improved by 14% (P less than 0.001) and 11%, respectively, in the exercise group, whereas there were corresponding reductions of 6 and 8% in the control subjects, with little change in fatigue index in either group. The similar relative increases in isokinetic peak power and peak VO2 suggest that improvement in short-term exercise capacity may be an important contributor to the improvement in aerobic exercise tolerance frequently observed in cardiac patients undergoing an endurance exercise program.     

Validity and reliability of the PowerTap mobile cycling powermeter when compared with the SRM Device

The SRM power measuring crank system is nowadays a popular device for cycling power output (PO) measurements in the field and in laboratories. The PowerTap (CycleOps, Madison, USA) is a more recent and less well-known device that allows mobile PO measurements of cycling via the rear wheel hub. The aim of this study is to test the validity and reliability of the PowerTap by comparing it with the most accurate (i.e. the scientific model) of the SRM system. The validity of the PowerTap is tested during i) sub-maximal incremental intensities (ranging from 100 to 420 W) on a treadmill with different pedalling cadences (45 to 120 rpm) and cycling positions (standing and seated) on different grades, ii) a continuous sub-maximal intensity lasting 30 min, iii) a maximal intensity (8-s sprint), and iiii) real road cycling. The reliability is assessed by repeating ten times the sub-maximal incremental and continuous tests. The results show a good validity of the PowerTap during sub-maximal intensities between 100 and 450 W (mean PO difference -1.2 +/- 1.3 %) when it is compared to the scientific SRM model, but less validity for the maximal PO during sprint exercise, where the validity appears to depend on the gear ratio. The reliability of the PowerTap during the sub-maximal intensities is similar to the scientific SRM model (the coefficient of variation is respectively 0.9 to 2.9 % and 0.7 to 2.1 % for PowerTap and SRM). The PowerTap must be considered as a suitable device for PO measurements during sub-maximal real road cycling and in sub-maximal laboratory tests.